Tumor-derived small extracellular vesicles loaded with functionally active miR-34a mimic can modify the anti-tumor response in 4T1 breast cancer animal model

Abstract

Background

A highly detectable problem in Triple-negative breast cancer is reduction of tumor suppressor miRNAs. Thus, this study aimed to use tumor-derived small extracellular vesicles (tsEV) obtained from 4T1 cells as a vehicle for miR-34a-replacement therapy (tsEV-miR-34a-mimic).

Methods

We tested 4T1-tsEVs freely, loaded with miR-34a-mimic or miR-34a-inhibitor on 4T1-bearing mice. Frequency of T cells in spleen and Inguinal lymph nodes assessed by flow cytometry then the relative gene expression of target genes of miR-34a evaluated by Real-Time PCR. In addition, level of cytokine secretion considered by ELISA. Additionally, MTT and AnnexinV/PI methods were used to investigate treatments on 4T1 cell proliferation and apoptosis rate, respectively. Afterwards, histopathological evaluation is applied to determine the extent of metastasis. Ultimately, in each group, mice were followed up on to assess the effect of treatments on survival rate.

Results

Treatment with tsEV-miR-34a-mimic profoundly increased survival and reduced metastasis in 4T1-bearing mice compared to other groups. Besides, the frequency of CD8 T cells was amplified in tumor tissue and inguinal draining lymph nodes (IDLNs). CD4T cell’s polarization toward regulatory T cells (Treg) was reduced. Gene expression pattern of tumor tissue showed a change from immune-suppressive to immune-activating tumor-microenvironment (TME). IL-6 and TGF-β concentrations were significantly reduced. IDLN lymphocytes performed a robust killing ability and actively proliferated in response to 4T1-lysate.

Conclusion

Totally, tsEV could be considered as a delivery carrier for miR-34a replacement therapy which also provided an anti-tumor immune response against 4T1- tumor. Thus, this platform could be considered a complementary approach to TNBC therapy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12935-025-03994-6.

Background

On global scale, among women, breast cancer still takes the first place for being the most prevalent type of cancer. According to the American Cancer Society estimations, from 2014 to 2018 an annual 0.5% increase is detectable for the incidence of breast cancer which lead to almost one-third of cancer cases in women [1]. Breast cancer accounts as a heterogenous disease and 15–20% of diagnosed breast cancer cases are Triple Negative Breast Cancer (TNBC). TNBC unlike other types of breast cancer lacks any specific cancer antigen, which is clinically recognized for its poor prognosis, high metastatic behavior, and low immunogenicity. Even though, surgery and chemotherapies along with new combination therapies provide hopeful outcomes in TNBC, nonetheless, finding new therapeutic platforms is still ongoing [2].

MicroRNA (miR)- Replacement Therapy (MRT), is one of the new therapeutic approaches that can be a supplement to common cancer therapies. In the context of cancer, miRs, a class of small noncoding RNAs, generally act as either tumor suppressors (tsmiR) or are known as oncomiRs, that amplify tumor progression [3]. MiR-34a is one of the well-known tsmiRs [4]that is reported to be more than three times decreased in human TNBC cell lines [5]. Moreover, lower miR34a levels in tumor tissues have shown a correlation with higher metastasis and lower survival rate [6]. Corresponding to different studies, miR-34a suppresses different aspects of breast cancer progression including angiogenesis [7], metastasis [8] and proliferation [9, 10]. Therefore, miR-34a become the first candidate for heading to clinical trial in cancer therapy [11].

One of the pillars of the tumor microenvironment (TME) are immune cells. Interleukin-6 (IL-6) is one of the leaders of chronic inflammation in different cancer types which paves the way for cancer-related immune exhaustion/suppression [12]. Accordingly, the anti-cancer effects of miR-34a are not limited to cancer cells and it has been demonstrated that miR-34a could manipulate the immune system by targeting IL-6/Signal Transducer and Activator of Transcription3 (STAT3) and reducing differentiation of CD4 T cells to Th17 in colorectal cancer (CRC) [13]. Adding to this, Dequan Wu et al. [14] and Huang et al. [15] separately declared that miR-34a targets SMAD4 which ultimately suppresses Tumor Growth Factor -β (TGF-β). In another study on Acute Myeloid Leukemia (AML), Programmed Death-Ligand 1 (PD-L1) was introduced as a direct target of miR-34a which nominated the miR-34a mimic as a potential immunotherapeutic agent [16]. Recently, a study showed that miR-34a directly targeted Fork-Head Box P3 (FOXP3) and reduced the regulatory T (Treg) population [17]. Despite these beneficial effects, a great barrier in MRT is providing the perfect carrier.

Different nanocarriers were applied for TNBC during past years but despite some success, research for finding the perfect carrier is still ongoing. New remarkable nanocarriers are extracellular vehicles (EVs) that are lipid bilayer cell-made sacs with different sizes. Small EVs (sEVs) also known as exosomes, are 30–200 nm in size and take part in cellular homeostasis and communication [18]. What makes sEVs more attractive than other nanocarriers are their low immunogenicity/toxicity, biocompatibility, stability in circulation, and bio-barrier permeability [19]. Also, cellular stress such as starvation [20] accelerate sEV production. Like other cells, tumors produced sEVs (tsEV) that not only have the above-mentioned features but also carry tumor-associated antigens (TAAs), major histocompatibility complex (MHC) residues, and some co-stimulatory parts which empowered them to stimulate immune cells and amplify the anti-tumor immune response [21–24]. Of interest, tsEVs have been shown to preferentially home to their parent cancer cell and stay longer in tumor tissue, which provides an intrinsic selective delivery aspect for them [25].

Since, TNBC, intrinsically is an immunologically cold tumor [26] with poor prognosis in individuals with low Tumor Infiltrating Leukocytes (TIL) and immunosuppressive atmosphere [27, 28], we took advantage of miR-34a loading in tsEVs (tsEV-miR) as a therapeutic approach for 4T1 breast cancer model to inhibit cancer related biological pathways in addition to induction of anti-tumor immune response.

Methods

Cell culture & TsEV isolation

4T1(metastatic adherent epithelial murine breast cancer cells) and CT-26 cell lines (colorectal carcinoma cell line with murine origin, highly tumorigenic and with an adherent fibroblast-like morphology) were purchased from Pasteur Institute of Iran, Tehran, and maintained in RPMI1640 (Bioidea, Tehran, Iran) containing 10% fetal bovine serum (FBS, Gibco, UK), 100u/ml penicillin and 100 µg/ml streptomycin at 37 °C in a humidified 5% CO₂ atmosphere. As cells reached a 70–80% confluency, media was removed. Cells were washed with phosphate buffer saline (PBS) two times, afterwards cell culture continued by adding serum-free media for another 24 h. The next day conditioned media (CM) was collected.

Isolation of tsEVs was done by an Exocib kit (Cibzist, Tehran, Iran) according to manufacturer protocol. First fresh CM was subjected to two-step centrifuges at 300 g for 5 min and 2000 g for 10 min to remove any cellular debris. For better purification, the supernatant was filtered by 0.22 μm filters. Filtered CM was added to a polyethylene glycol (PEG) solution in a ratio of 5:1 and after 5 min vortexing, subjected to 4 °C overnight incubating. The day after that, the prepared solution was centrifuged at 800 g for 40 min. Next, the tsEV-enriched remaining pellet was aliquoted and kept at −80 °C [29].

Characterization of isolated TsEVs

The size characterization of isolated tsEVs was carried out by dynamic light scattering (DLS, ZetaSizer Nano series Nano-ZS, Malvern Instruments Ltd., Malvern, UK). Protein concentration of purified tsEVs calculated by Bicinchoninic acid assay (BCA, Thermo Scientific, Rockford, USA). Morphology of collected tsEVs was examined by field emission scanning electron microscope (FESEM, S4800, Hitachi Co., Japan) and transmission electron microscope (TEM), respectively [29].

Loading TsEVs with miR-34a mimic and miR-34a inhibitor

miR-34a-mimic and inhibitor (Bioneer, Korea) were separately loaded in tsEVs based on the modified CaCl₂ method [30]. In detail, 150 µl of tsEV with a concentration of 1 µg/µl (based on protein concentration results from BCA assay) was added to 150 µl CaCl₂ (0.1 M), and ultimately, 1 µl of mimic or inhibitor (100pmol) was added to the solution and mixed well. To transfect, the solution was placed on ice for 30 min, transported to 42 °C for 60 s, and as a final step, again incubated on ice for 5 min. Next, to degrade any unloaded miR molecule, RNase A (EN0531, Thermo Fisher Waltham, MA) was added to the prepared tsEV-miR solution which was accompanied by the addition of RNase-inhibitor (Thermo Fisher) to stop the reaction, according to manufacturer’s instructions. Afterwards, the mixture was re-purified with an EV-isolation kit [19].

Cellular uptake of tsEV-miR34a

4T1 cells were cultured in 24 wells plate at 2.5 × 10⁵ density. The next day, cells were treated with 25 µg/ml of tsEV-miR-34a-mimic and tsEV-miR-34a- inhibitor. Untreated cells were used as the control group. After 24 h, the media was removed and the cells were washed with PBS, afterwards RNA was extracted from the cells of each well with exoRNeasy Midi Kit (Qiagen, Germany). The cDNA synthesis process was done with a Biofact (Korea) cDNA synthesis kit. Sybr-green-real-time PCR was assigned for relative expression of miR-34a- mimic and inhibitor. U6 expression was used as the internal control [19].

4T1 mouse model

Female BALB/c mice 6–8 weeks old were purchased from Royan Institute (Tehran, Iran) and were housed in the animal research facility. Mice were kept under the standard 12 light/dark cycle at 22–24 °C and provided with water and food ad libitum. All animal protocols of this study were approved by the Institutional Ethical Committee and Research Advisory Committee of Shahid Beheshti University of Medical Sciences (ID: IR.SBMU.MSP.REC.1398.581).

4T1-mouse model induction was performed by direct subcutaneous (s.c.) injection of 5 × 10⁵ 4T1- cells in 50 µl phosphate buffer saline (PBS) to the pre-shaved right flank of the mice [31]. Tumor growth was screened and after 10 days, when the tumors reached an approximate size of 100 mm³, mice were randomly divided into 5 groups (n = 8). The treatments were as follows: PBS, CT-26-tsEV as negative control tsEV (NC-tsEV) for 4T1-tsEV, 4T1-tsEV (tsEV), tsEV-miR34a-mimic and tsEV-miR-34a-inhibitor. Each group received 3 peri-tumoral s.c. injection of the mentioned therapeutic agents (30 µg of total protein in 60 µl volume) at 3-day intervals. The PBS group got a 60 µl of PBS injection at the same site [32]. Tumor volumes were measured on a twice-weekly chart with Vernier Calipers according to the equation: V = Length× Width²/2. Seven days after the last treatment, three mice from each group were sacrificed for further ex-vivo tests and five mice were followed up for survival evaluation. When the tumor volume of the survival experiment mice reached 2000mm³, for ethical reasons, the mice were euthanized by cervical dislocation (Fig. 1).

Fig. 1.

An illustration of the study design, in vitro & in vivo tests used for evaluation of immune response after tsEV-miR-34a-mimic therapy

Hematoxylin and Eosin (H&E) tissue staining

To examine the metastasis rate, inguinal draining lymph node (IDLN), liver, and lungs were taken and for assessing the changes in tumor-infiltrating leukocytes (TIL) and necrosis, tumor tissues were gathered. All tissues were maintained in 10% buffered formalin and then paraffin-embedded. A 5 μm- thick cut of tissue sections were prepared and placed on glass slides. The slides were stained with hematoxylin and eosin and imaged under light microscopy [19].

Lymphocyte isolation & flow cytometry

Tumors, IDLNs, and spleens of sacrificed mice were used in the lymphocyte isolation process [19]. For this purpose, tumor tissues were washed with cold PBS, cut into pieces, and digested by collagenase type IV (10 mg/ml, Sigma-Aldrich) at 37 °C for 30 min. IDLNs and spleens were kept in RPMI1640 and minced softly by sterile scalpers. Each of the digested mixtures of tumor tissues and the suspensions obtained from IDLNs and spleens were passed through a 70 μm cell strainer before they were taken to centrifuge (4 °C, 300 g, 5 min). The cell pellets were exposed to 5 ml of ACK hypotonic lysis solution (Sigma, Seoul, Korea) for 5 min at room temperature (RT) to remove red blood cells (RBCs). Next, the isolated lymphocytes were washed twice with cold PBS (4 °C, 300 g, 5 min).

To evaluate the frequency of different T cell subtypes, 1 × 10⁶ isolated lymphocytes from each tissue were first suspended in 100 µl staining buffer (PBS + 2% FBS) and then subjected to surface marker staining by adding 1µL of anti-mouse antibodies for CD8 (FITC conjugated), CD4 (PE-conjugated) and CD3 (Percp-Vio700 conjugated) (Miltenyi Biotec, Gladbach, Germany) for 15 min at 4 ◦C in the dark. For assessing Tregs, 1 × 10⁶ cells were first stained with the surface marker antibodies including, anti-mouse CD4-FITC and anti-mouse CD25-PE (BD, USA) by the previously mentioned protocol. Afterward, the cells were washed and resuspended in 200 µl of perm-fix buffer (BD, USA) for 20 min at 4 ◦C in the dark, and then washed twice with 1 ml of perm/wash buffer (BD-USA). The pelleted cells were resuspended in 100 µl perm-wash buffer while 1 µl of anti-mouse Foxp3-APC conjugated (BD-USA) antibody was added. The cells were incubated for 30 min at 4 °C in the dark. Again, the cells were washed twice with 1 ml of perm-wash buffer and ultimately resuspended in 100 µl of staining buffer. For Treg evaluations, cells were stained with 1 µl of zombie NIR fixable Viability kit (Biolegend, USA) for 15 min at 4 °C in the dark and washed with PBS, before any other staining. To acquire fluorescent activated cell sorting (FACS) data, samples were read with a BD FACSCanto II (BD Biosciences), and the related data were analyzed with FlowJo v10 (TreeStar, Inc.).

Quantitative real-time PCR

Tumor tissue samples from each mouse were first snap-frozen, and 30 mg of the frozen tissues were cut into pieces on a sterile petri dish with a sterile scalper. Total RNA was extracted with a total RNA extraction kit (Parstous, Mashhad, Iran), and 1 µg of extracted RNA was used for cDNA synthesis with an easy cDNA synthesis kit (Parstous, Mashhad, Iran), according to the manufacturer’s instructions. To evaluate target gene expression in proportion to housekeeping gene Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), real-time PCR was used by applying 1 µl of cDNA, 0.5 µl from each of reverse and forward primers (10 pmol) with 7 µl of ExcelTaq™ 2X Fast Q-PCR Master Mix (SYBR, no ROX) (SMOBIO, South Korea) in a total volume of 14 µl. The primers (Sinaclone, Tehran, IRAN) are listed in Table 1 ([19, 33]) and were either designed by oligo v. 7.56 software (Molecular Biology Insights, Inc., CA, USA) and blasted with NCBI or adapted from previous studies. The obtained data were analyzed by the 2^−ΔΔct method.

Table 1.

Primer sequences used in this study

Gene	Primer sequence 5’−3’	Reference
GAPDH	F: AACGACCCCTTCATTGAC R: TCCACGACATACTCAGCAC	[23]
IL-10	F: GGACTTGAAGTGCCATTGGT R: CATCACGATCTCCCGGTTAT
IFN-γ	F: GCTTTAACAGCAGGCCAGAC R: GGAAGCACCAGGTGTCAAGT
TGF-β	F: TGGAGCAACATGTGGAACTC R: TGCCGTACAACTCCAGTGAC
IL-17	F: TCTCATCCAGCAAGAGATCC R: AGTTTGGGACCCCTTTACAC
IL-6	F: TAGTCCTTCCTACCCCAATTT R: TTGGTCCTTAGCCACTCCTT
MMP2	F: GCACACCAGGTGAAGGATGT R: GGTGAAGGAGAAGGCTGGTT
MMP9	F: TCTCCCGAGAGTCCAACTCA R: TGGATCAGTTCCAGCTGAGG
VEGF	F: TCGCTCCTCCACTTCTGAGG R: GGCCATTACCAGGCCTCTTC
Foxp3	F: TACACCCAGGAAAGACAGCAACCT R: TCTGCTTGGCAGTGCTTGAGAA
STAT3	F: TCGCTCACGTTTGACATGGA R: TCTAACAACCAACCCCGAGC	[17]
PDL1	F: ACTTGCTACGGGCGTTTAC R: CTGAAGTTGCTGTGCTGAGG
Rorc	F: TGAAGGCAAATACGGTGGTGTG R: CAGGACGGTTGGCATTGATGAG	This study
Tbx21	F: TCCAACAATGTGACCCAGATG R: GTTCTCCCGGAATCCTTTGG
GATA3	F: GCTCCTTGCTACTCAGGTGAT R: GGAGGGAGAGAGGAATCCGA
Mmu-miR34a-5p	F: 5′ -GCGGCGGTGGCAGTGTCTTAGC-3′ R: 5′ -ATCCAGTGCAGGGTCCGAGG-3′	[19]
SnRNAU6	F: 5′ -GCTTCGGCAGCACATATACTAAAAT-3′ R: 5′ -CGCTTCACGAATTTGCGTGTCAT-3′

GAPDH: glyceraldehyde-3-phosphate dehydrogenase, IL-10: Interleukine-10, IFN-γ: interferon-gamma, TGF-β: tumor growth factor-β, IL-17: interleukine-17, IL-6: interleukine-6, MMP2: matrix metalloproteinase 2, MMP9: matrix metalloproteinase 9, VEGF: vascular endothelial growth factor, Foxp3: forkhead box protein 3, STAT3: signal transducer and activator of transcription 3, PDL1: programmed death ligand 1, Rorc: RAR related orphan receptor c, Tbx21: T-box transcription factor 21, GATA3: GATA binding protein 3

Enzyme-linked immune sorbent assay (ELISA)

Isolated lymphocytes of IDLNs and spleens were seeded in 96 well plates at a density of 1 × 10⁶ cells/well in the presence of complete RPMI1640 media. To boost tumor-specific immune responses, cells were treated with 80 µg/ml of sterile 4T1 cells- lysate that was prepared by repeated freeze (−196 °C) and thaw (37 °C) cycles [34]. Untreated cells were used as a negative control. The supernatant of each well was gathered 48 h later and maintained at −80 °C for further evaluation of IL-10, IL-6, and TGF-β concentrations with Mouse ELISA MAX standard set (Biolegend, USA) according to the manufacturer’s protocols.

Apoptosis test

4T1 cells were cultured in 48 wells plate at a density of 2 × 10⁴, 12 h before conducting the cytotoxicity test. Isolated Lymphocytes from spleens and IDLNs were added to each mentioned well in a target: effector ratio of 1:10 (2 × 10⁵ cells). After 6 h of 4T1 lymphocyte coculture, the media of each well was aspirated and washed twice with PBS to remove any remaining suspended lymphocytes [19]. To detect lymphocyte-induced apoptosis percentages in 4T1 cells, a Biolegend FITC-Annexin V/Propidium Iodide (PI) Apoptosis Detection Kit (San Diego, CA) was used. First 4T1 harvested cells were resuspended in Annexin V binding buffer, next the cells were stained with 5µl Annexin V and 10 µl PI. The stained cells were incubated at RT for 20 min in the dark. Ultimately, the cells were washed, resuspended in 400 µl PBS, and were read by BD FACS Canto flow cytometry, the received data were analyzed using FlowJo 7.6.1 software (Tree Star, Ashland, OR, USA). Non-co-cultured 4T1 cells were considered as negative controls.

Lymphocyte proliferation assay

To evaluate IDLNs and spleens-derived lymphocyte proliferation capacity towards tumor antigens, an MTT assay was performed as previously described [35]. In brief, the isolated lymphocytes were seeded at a density of 5 × 10⁵ cells in 96 wells plates 12 h before test initiation.

Afterward, the cells were either treated with 4T1-derived Lysate at the concentration of 80 µg/ml, poly hydroxy alkonate (PHA) (Gibco, USA) at 2% v/v, or left untreated as a negative control. After 48 h, 20 µl of MTT solution (5 mg/ml) (Sigma, USA) was added to each well, 4 h later, cells were exposed to 100 µl dimethyl sulfoxide (DMSO). Each well-related absorbance was measured by an ELISA reader at 570 and 630 nm (reference). To calculate the stimulation index by Optical Density (OD) data the following equation was used:

Statistical analysis

The data were represented as the means ± standard error of the mean (S.E.M). One-way/two-way ANOVA or Mann–Whitney U-test was used to perform a comparison between groups. Kaplan-Meier method was used for survival evaluations. The GraphPad Prism 8 (GraphPad Software, Inc., La Jolla, CA, USA) was applied for drawing graphs. A P < 0.05 was considered as statistically significant.

Results

Characterization of 4T1- derived TsEVs & loading with miR-34a

As shown in Figs. 2a and 4T1-derived tsEVs size was 130.9 ± 10.99 nm. Their cup shaped and spherical morphology was confirmed by TEM and FESEM images, respectively (Fig. 2b, c). According to the BSA results, 2800 µg/ml of tsEV was obtained from 150 ml of FBS-free media supernatants of 4T1 cultured cells. The relative expression of miR-34a in tsEV-miR-34a-mimic group was 96.53 ± 0.796 whereas its expression was 2.03 ± 0.145 and 0.108 ± 0.044 in tsEV and tsEV-miR-34a-inhibitor, respectively, compared to control group (1 ± 0.03) (Fig. 2d).

Fig. 2.

tsEV characterization and its loading with miR-34a. (a) tsEV average size measured by DLS (130 ± 10.99 nm). (b) Transmission electron microscopy (TEM) image of isolated tsEVs pointed by red arrows. (c) Field emission scanning electron microscopy image (FESEM) of isolated tsEVs. (d) Relative expression of miR-34a in 4T1 cells after treatment with tsEV-miR-34a-mimic, compared to tsEV, tsEV-miR-34a-inhibitor and non-treated cells (one way ANOVA, means ± S.E.M, *P < 0.05, ***P < 0.001)

tsEV-miR-34a-mimic increase CD8T frequency in tumor and IDLNs

As represented in Fig. 3, the frequency of CD8T cells in tumor tissue of the tsEV-miR-34a-mimic treated group (14.5 ± 0.058) was significantly higher than other treatments (p < 0.001) and control group (3.487 ± 0.185, P < 0.001). CD4T cells population showed an increased trend in all treatments against the control group (2.663 ± 0.118), the highest increase was related to mimic (10.617 ± 0.802, P < 0.001) and tsEV treated groups (8.663 ± 0.5, P < 0.001), respectively. There were no differences between NC-tsEV and inhibitor treatments (P > 0.05). However, no statistically significant changes were detectable in CD4/CD8 ratio between PBS and mimic group, TCD4/CD8 proportion in mimic received group was dramatically lower than other types of therapies (NC-tsEV: 1.17 ± 0.039, P < 0.01, tsEV: 1.677 ± 0.11, P < 0.001, tsEV-miR-34a-inhibitor: 1.57 ± 0.01, P < 0.001).

Fig. 3.

CD4T and CD8T frequency and their ratio in tumor, IDLN, and spleen of different treatments. a. The frequency of CD4T, and CD8 T cells in the tsEV-miR-34a-mimic group compared to tsEV, NC-tsEV, tsEV-miR-34a-inhibitor, and the control group by Two-way ANOVA showed a pattern of increasing in CD8T cells in tumor and IDLN but not spleen. b. CD4 T/CD8T ratio analysis by one-way ANOVA in tsEV-miR-34a-mimic group compared to tsEV, NC-tsEV, tsEV-miR-34a-inhibitor and control group revealed a decline in tsEV-miR-34a mimic therapy (n = 3, means ± S.E.M *P < 0.05, **P < 0.01, ***P < 0.001,, ns, non significant)

In IDLNs, CD8T cells were increased in the mimic group (12.53 ± 0.318) compared to PBS (10.76 ± 0.518, P < 0.05) and inhibitor (9.517 ± 0.356, P < 0.001) received groups. CD4T cells exhibited no changes among PBS and mimic therapy (P > 0.05), and in opposition to this, other therapeutic agents displayed a higher frequency toward control group. In Spleen tissue, neither CD4 T nor CD8T cells revealed any specific changes in frequency (P > 0.05). Of note, the gating strategy was depicted as supplementary file 2. Dot plot data of CD3/CD4 and CD3/CD8 of each group was presented in supplementary file 3 and 4.

TsEV-miR-34a-mimic reduced T regulatory population in Tumor, IDLN, and spleen

Evaluation of the Treg population in tumor tissue showed a sharp decrease in the mimic-treated group against control (2.86 ± 0.411, P < 0.05) and inhibitor (2.64 ± 0.498, P < 0.05). No specific changes were detectable between mimic and other groups(P > 0.05). Besides, in IDLNs, Treg pictured a strong decline in mimic-therapy toward control (13.13 ± 0.20, P < 0.001), and other tested groups. In the spleen, the Treg population was reduced in NC-tsEV, tsEV, and inhibitor compared with the control group (7.23 ± 0.65, P < 0.001) but the sharpest reduction was dedicated to mimic-therapy (0.275 ± 0.025, P < 0.001) (Fig. 4). Of interest, dot plot data of each group are accessible in supplementary file 5.

Fig. 4.

Treg frequency and its ratio in proportion to CD8T cells in different tissues after treatment. (a) Bar plots of comparisons of evaluated Treg frequencies after different therapies. The population of Treg fell in the tumor, IDLN, and spleen by tsEV-miR-34a-mimic therapy. (b) Treg/CD8T ratios in different tissues of tested subjects. The highest decline was seen in the Treg/CD8T ratio of tsEV-miR-34a-mimic therapy which reflected a major CD8T rise and Treg drop by this treatment toward other therapies. The data was shown as means ± SEM, (n = 3,*P < 0.05, **P < 0.01, ***P< 0.001, ns, non significant) and analyzed by one-way ANOVA

TsEV-miR-34a mimic modified Immune-Evasion and cancer promoting gene expression profile

Based on tumor tissue gene expression data (Fig. 5), in tsEV-miR-34a mimic (0.4407 ± 0.0115, p < 0.001) and to a lesser extent in tsEV treated groups (0.7117 ± 0.005, P < 0.001), STAT3 was efficiently down-regulated in comparison to PBS receiving group (1.027 ± 0.014). In NC-tsEV and tsEV-miR-34a-inhibitor groups, the STAT3 was over-expressed (P < 0.001).

Fig. 5.

Relative expression of tumor-promoting and immune-inhibitory targets of miR-34a in tumor tissue. In total a sharp drop was observed in tsEV-miR-34a-mimic group toward other treated groups in all evaluated genes except for MMP2. The comparison was made between control, NC-tsEV, tsEV, tsEV-miR-34a-mimic/inhibitor by one-way ANOVA and represented as means ± SEM (n = 3,*P < 0.05, **P < 0.01, ***P < 0.001, ns, non significant)

PD-L1 showed a pattern of expression almost like STAT3. In tsEV-miR-34a mimic (0.262 ± 0.26), PD-L1 was highly decreased and in tsEV group (0.44 ± 0.020) a slight decrease in expression was detected in proportion to PBS (1.027 ± 0.014, P < 0.001).

For VEGF, only tsEV-miR-34a-mimic was capable of inducing a significant reduction of expression toward the PBS group (0.539 ± 0.0125 vs. 1.023 ± 0.0145, P < 0.05).

The expression of MMP9 was dramatically dropped in tsEV-miR-34a-mimic (0.215 ± 0.010) group across other treatments (P < 0.001). No significant changes were detectable in the expression of MMP2 in mimic group toward other groups except for NC-tsEV (P > 0.05).

TsEV-miR-34a mimic modulate tumor cytokine expression profile

The tumor cytokine pattern of expression showed a rotation toward immune-promoting cytokines (Fig. 6). IL-6 expression was mainly fell by tsEV-miR-34a-mimic (0.622 ± 0.029, P < 0.001), and to a lesser degree by tsEV (0.779 ± 0.1014, P < 0.001) treatments toward tsEV-miR-34a-inhibitor. Although all treatments caused an increase of IFN-γ expression but a sharp rise of IFN-γ expression was detectable in tsEV-miR-34a-mimic group compared with PBS group (5.15 ± 0.057 vs. 1 ± 0.030, P < 0.001). IL-17 relative expression demonstrated a big jump in tsEV-miR-34a- mimic (1.193 ± 0.034) across other groups especially tsEV (0.163 ± 0.046, P < 0.001).

Fig. 6.

Relative expression of cytokine genes in tumor tissue. Changes in tumor cytokine profile regarding different therapeutic regimens were analyzed by one-way ANOVA and displayed as means ± SEM. (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ns, non significant)

Through the tested groups TGF-β expression showed a relative decline against control group and the highest decline rate goes to tsEV-therapy (0.186 ± 0.007, P < 0.001). IL-10 relative expression decreased in all treatments (P < 0.001) compared to control but tsEV-miR-34a-mimic (0.11 ± 0.013, P < 0.001) drastically reduced its expression. On the other side, aside from tsEV-miR-34a-mimic (0.642 ± 0.008, P < 0.001), the expression of IL-4 in all treatments were significantly higher than control group (P < 0.001).

TsEV-miR-34a-mimic induced Thelper1(Th1) cells population

To evaluate changes in CD4T subtypes in TME, the relative expression of each CD4 T master transcription factor was detected. TBX relative expression showed an increase only in tsEV-miR-34a-mimic therapy (1.11 ± 0.049) and the lowest amount was detected in tsEV (0.355 ± 0.023) treated group compared to other groups (P < 0.001).

Although, all tested therapies reduced RORγc expression compared to the control group (PBS: 1 ± 0.014, P < 0.05)) but the highest reduction was related to tsEV-miR-34a-mimic (0.113 ± 0.046, P < 0.001).

Foxp3 expression in tumor tissues resembled the RORγc pattern except that no changes were shown between NC-tsEV and tsEV-miR-34a-inhibitor (P > 0.05). Reduction of Foxp3 expression was exacerbated in tsEV-miR-34a-mimic group compared to tsEV treated samples and control group (1 ± 0.025, P < 0.001).

Despite great drop of GATA3 in tsEV and tsEV-miR-34a-mimic groups against control (1 ± 0.0185) and NC-tsEV (0.691 ± 0.0049, P < 0.001), but these treatments displayed no specific changes with tsEV-miR-34a-inhibitor (P > 0.05) (Fig. 7).

Fig. 7.

T cells master transcription factors relative expression in tumor tissue. Differences in transcription factors of T cells main lineages in the tumor after tsEV-miR-34a-mimic therapy with domination of Th1 response exhibited as mean ± SEM and analyzed by one-way ANOVA (n = 3, *P < 0.05, **P < 0.01 ***P < 0.001,ns, non significant)

TsEV-miR-34a-mimic skewed cytokine profile toward immune activation

According to our flow cytometry and cytotoxicity data, IDLNs seemed a better mirror than spleen for evaluating changes in tumor response, thus we investigated the miR-34a-related cytokine profile in the supernatant of lymphocytes derived from IDLN after exposure to 4T1- lysate. The concentration of IL-6 illustrated a specific, ordered and sequential decrease from tsEV-miR-34a-inhibitor (65.52 ± 2.795, P < 0.001) to tsEV (44.58 ± 2.70, P < 0.001) and the lowest amount was related to the mimic injected group (30.26 ± 2.45, P < 0.001), all compared to the control group (95.44 ± 3.24).

Contrary to the observed pattern for IL-6, IL-10 was released in high levels from lymphocytes of mimic-treated mice (3168 ± 68.21, P < 0.001) to tsEV (2825 ± 35.36, P < 0.001). TGF-β dramatically reduced in tsEV-miR-34a-mimic therapy compared to tsEV-miR34a-inhibitor, NC-tsEV and tsEV (P < 0.01). (Fig. 8).

Fig. 8.

ELISA assay for IDLNs. After 48 h of treatment with 4T1 lysate, the concentration of cytokines in IDLN lymphocytes was evaluated by one-way ANOVA and showed as mean ± SEM (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ns, non significant)

TsEV-miR-34a- mimic Amplified, tumor apoptosis ability of IDLNs lymphocytes

After co-culture of IDLNs lymphocytes with 4T1 cells, the AnnexinV/PI flow cytometry results exhibited that the tsEV-miR-34a-mimic (20.94 ± 0.521) induced a powerful apoptosis rate in 4T1cells in comparison to control (10.78 ± 0.385), NC-tsEV, tsEV and tsEV-miR34a-inhibitor (P < 0.001). A meaningful apoptosis progression was also seen for the tsEV received group across control (P < 0.001), NC-tsEV (P < 0.01), and inhibitor-treated groups (P < 0.001). No changes were observable between the PBS and inhibitor groups. Evaluating splenocytes for 4T1 cell killing ability showed no discrepancies between different studied groups (P > 0.05) (Fig. 9a-b). Of note dot blot data of apoptosis has been depicted in supplementary file 6.

Fig. 9.

Killing capacity and stimulating index of IDLN and spleen isolated lymphocytes.(a) Comparison of percentages of 4T1 apoptosis in tsEV-miR-34a-mimic treated mice across other treatments by IDLN lymphocytes with one-way ANOVA test. (b) 4T1 cells apoptosis percentage after exposure to splenocytes analyzed by one way ANOVA.c. Rate of proliferation in Lymphocytes of different treated groups toward control after stimulation with 4T1 cell lysate analyzed by two-way ANOVA test. (d). Rate of proliferation in splenocytes of different treated groups toward control after stimulation with 4T1 cell lysate. (Two-way ANOVA test) The data was represented as mean ± SEM (n = 3, *P < 0.05, **P < 0.01, ***P< 0.001, ns, non significant)

TsEV-miR-34a-mimic boosted IDLNs specific proliferation capability against 4T1-Lysate

We have used MTT assay and imaged our proliferation results in Fig. 9c & d. The calculated stimulation index (SI), manifested a strengthened proliferative ability in tsEV-miR-34a-mimic (2.728 ± 0.057) received group in contrast to PBS (0.892 ± 0.039), NC-tsEV, tsEV and tsEV-miR-34a-inhibitor (P < 0.0001).

Histopathological findings

According to the tumor tissue H&E staining results, all groups had grade three tumors. No differences were seen in the rate of nuclear pleomorphism and mitosis. The total percentages of TIL accumulation were increased in treated groups especially mimic injected mice compared to control (Fig. 10e). Apoptosis and extensive necrosis were seen in tsEV-miR-34a-mimic compared to tsEV-miR-34a-inhibtor and the control group (Fig. 10d). Metastasis was identified in the lung, and liver, besides, full metastasis was detected in tumor-draining lymph nodes of the PBS group. In contrast, tsEV-miR-34a-mimc group lung, liver, and draining lymph node were free of any metastasis. A mild metastasis was seen in the lung and liver of the tsEV-miR-34a inhibitor-treated group (Fig. 10a-c).

Fig. 10.

Histopathological findings. Metastasis in IDLN (a), lung (b), and liver(c) of different therapies. In tsEV-miR-34a-mimic treated mice all evaluated tissues were metastasis-free compared to PBS received group. (d) necrosis of tumor tissue which was bordered by dotted lines and black stars. The highest rate of necrosis was detected in the tsEV-miR-34a-mimc group compared to other therapeutic agents. (e) Tumor infiltrating lymphocytes after different treatments were shown by red arrows (x400). An increase in the accumulation of stromal TIL was seen in the tsEV-miR-34a-mimic-treated group. (Scale bar x100: 1600 μm, Scale bar x400: 400 μm, A: PBS, B: NC-tsEV, C: tsEV, D: tsEV-miR-34a-mimic, E: tsEV-miR-34a-inhibitor)

TsEV-miR-34a extended survival and reduced tumor size and weight

Tumor weight and size are displayed in Fig. 11. In all treatments the tumor size was significantly reduced in comparison to PBS (1.454 ± 0.095, P < 0.05) group. Tumor volume halt in progression was first started in tsEV-miR-34a (334.37 ± 56.27) treated group at day 19 compared to PBS (515.5 ± 36.5, P < 0.01). In day 21, NC-tsEV, tsEV and tsEV-miR-34a-inhibitor(P < 0.001) showed smaller volume against PBS group (906.93 ± 58). In addition, on day 21, mimic therapy (300.46 ± 43.77) could significantly reduce tumor volume in proportion to tsEV (P < 0.05) and tsEV-miR-34a- inhibitor (P < 0.001) groups. As depicted in Fig. 11, the median survival (ms) of each group were as follows: PBS 41 days (d), NC-tsEV:46 d, tsEV:61d, tsEV-miR-34a-mimic:63 d and tsEV-miR-34a-inhibitor: 45 d. A significant increase (P < 0.01) in ms was detected in mimic-treated group compared to PBS, NC-tsEV, and inhibitor-injected groups.

Fig. 11.

The anti-tumor effects of tsEV-miR-34a-mimic in 4T1 animal model. (a) Tumor volume was decreased significantly in the tsEV-miR-34a-mimic treated group compared to PBS receiving group (b) In all treated groups tumor weight was reduced drastically toward control. The maximum decline was caused by the tsEV-miR-34a-mimic group. c.tsEV-miR-34a-mimic treatment induced an increase in survival days of treated mice against PBS administered group. The data was represented as mean ± SEM and analyzed by Kaplan-Meier method (n = 5, *P < 0.05, **P < 0.01, ***P < 0.001, ns, non significant)

Discussion

TNBC is known as an aggressive tumor which causes many current therapies to fail. To find new therapeutic agents, we previously showed that tsEV-miR-34a could effectively down-regulate cancer-progressing genes in 4T1-treated cells. Moreover, we explained that by miR-34a mimic therapy, the 4T1 cell cycle was arrested and entered the apoptosis process which also caused proliferation suppression. In addition, miR-34a replacement therapy inhibited 4T1 cell migration ability. As an interesting finding, we exhibited that 4T1-derived tsEV could somewhat restrict 4T1 cells in terms of tumor-related gene expression and migration capability [29]. To translate our in vitro results to in vivo, in this study we evaluated the tumor-suppressing and immune-boosting benefits of tsEV-miR-34a-mimic therapy in 4T1 BALB/c bearing mice.

An essential feature in TNBC prognosis [36], survival [37], and response to immune therapy [38] is the frequency and function of TILs in TME. In our experiment, CD4 TILs has risen in all treatments in contrast for CD8 TILs, just tsEV-miR-34a-mimic group showed an increase toward control. In IDLNs, the same pattern was observed for CD8 T cells but a decrease was detected in CD4 T cells toward the rest of the groups. Moreover, evaluation of the Treg population in tumor tissue, IDLNs and spleen showed a sharp dropdown in Treg cells.

From a functional perspective, IDLN lymphocytes of the tsEV-miR-34a-mimic treated group manifested a vigorous killing ability by induction of 4T1 cells apoptosis compared to any other group whereas splenocytes lack this capability. Adding to this, tsEV-miR-34a expressed a tumor-specific pattern of proliferation which was missed in splenocytes. Evaluation of tumor tissue for CD4 related master transcription factors showed tsEV-miR-34a therapy could effectively down-regulate RORγc and FOXP3 where as TBX expression up-regulated against other therapies. This pattern was only detected for GATA3 compared to control and NC-tsEV groups.

In parallel to our study, it has been declared that in TNBC patients, high CD8 and CD4 scores is correlated with enhanced IFN-γ production, higher infiltration of other anti-cancer immune-related cells, and better overall survival (OS) [36]. In another opposing study, it has been claimed that higher tumor CD4 T cells led to poor prognosis in TNBC patients [39].

In describing the importance of Treg in the TNBC behaviour, Azogui et al. have announced that in the 4T1 model, depletion of Treg cells supplied an increase in the dendritic cells (DC) population which ultimately provided an effector PD-1^low CD8T cells [40]. In another study, they explained that massive accumulation of Treg cells in the 4T1 model reduced Th cell’s correct activation which subsequently caused a great loss in CD8 effector T cells [41]. Consistent with our observation, You et al. revealed that in the 4T1- model, Treg annulled specialized TCD8 anti-tumor immune responses in tumor-draining lymph nodes (TDLN) [42]. Therefore, according to our results and mentioned studies tsEV-miR34a-mimic therapy could mainly manipulate frequency and function of TNBC TILs, IDLN T lymphocytes, and to a lesser degree splenocytes by reducing Treg cells, inducing CD8T cells and changing Th cells subpopulations.

According to different studies, STAT3 acts as a crossroad manager molecule that co-operates with VEGF, MMP2, MMP9, and PD-L1 in different cancer-promoting processes [43].

Considering the report by Wai-Shiau-Chung, human TNBC cells use the SHP-1/p-STAT3/VEGF-A axis as a migration tool, and suppressing VEGF-related pathways could be beneficial action in controlling metastasis [44]. In addition, Niu et al. have explored the importance of STAT3 overexpression for MMP9 activation in SKBR3/EPR invasiveness [45]. In another study, Foukakis et al. reported that inhibition of STAT3 expression by pharmacologic agents or silencing methods, resulted in down-regulation of PD-L1, reducing tumor growth and metastasis, in vitro and in vivo, respectively [46].

From a clinical point of view, high expression of VEGF was related to poor disease-free and distant metastasis-free survival rate [44]. Besides, in some studies, higher levels of MMP9 were correlated with worse prognosis and survival of breast cancer patients [47, 48]. TNBC patients show higher levels of PDL1 compared to non-TNBC forms which was also correlated with poor OS [46].

Here, in accordance with the mentioned studies above, we represented that tsEV-miR-34a-mimic therapy could perfectly downregulate its direct target genes, STAT3 and PD-L1, compared to control group which contributes to downstream control of VEGF and MMP9 but not MMP2. In case of MMP2 the reduction in mimic receiving group did not show a synergistic pattern like other downstream targets which might be due to existence of other important regulatory molecules with more powerful effect, which needs further investigation to be clarified.

We have shown that, following STAT3 down-regulation in the mimic received group, IL-6 expression was dropped, too. In this regard, Brown et al. have shown that autocrine- IL-6 production is responsible for TNBC cell growth [49]. Adding to this, in another study, exposing THP-1 cells to miR-34a resulted in controlling the IL-6/IL-6R pathway that led to M1-cancer inhibiting phenotype [50].

TsEV-mir-34a- mimic therapy caused a decreasing pattern in other cancer-promoting cytokines as well as IL-6. Arteaga et al. have reported that applying TGF-β inhibitors amplified chemotherapeutic agents’ effect in TNBC [51]. By pointing to the approved role of miR-34a in the inhibition of Smad4, as the downstream target molecule of TGF-β in different cancers, our finding in harmony with previous reports, showed that TGF-β in tsEV-miR-34a receiving group is down-regulated in proportion to control group. As shown in different studies, other important cytokines in breast cancer are IL-4 [52] and IL-10 [53], which were fortunately fully decreased in the tsEV-miR-34a-mimic treated group compared to other groups.

Therapy with tsEV-miR34a-mimic not only reduced cancer-promoting cytokines but also over-expressed IFN-γ expression in tumor tissue. Based on the Purcell et al. study, IFN-γ treatment of TNBC cells caused positive changes in immunopeptidome by creating diversity in tumor antigen processing and presentation [54]. Moreover, related to different studies, IFN-γ has valuable anti-tumor/immune-activating effects in different cancers [55].

In our study, IL-17 was upregulated in tumor tissue in opposition to our expectations, which drew our attention to the dual role of IL-17 in cancer immunity [56]. Of note, low expression of STAT3 and RORγc are contradictory with IL-17 over-expression in tumor tissue. A possible explanation for this phenomenon could be newly found STAT3-independent IL-17 production by immune cells other than Th17 [57] and existence of intermediate Th17 cells with the ability to produce both IFNγ and IL-17 which could ultimately change to Ex-Th17-Th1- cells under pathologic conditions [58]. The mentioned changes require the existence of IL-12, IL-23, IL-1β, and a low amount of TGF-β with the cooperation of STAT4. During this identity change, an imbalance in transcription factors is inevitable [59]. Although we have detected reduced TGF-β expression as one of the pointed prerequisites, the exact relying mechanisms need further investigation.

Depending on the IDLN pattern of cytokine production after induction with tumor-lysate, TGF-β and IL-6 concentration supported our data obtained from tumor tissue which could be related to lower Treg and effective activation of lymphocytes in IDLN regarding tsEV-miR-34a treatment across other groups.

A conflicting finding was dedicated to IL-10 concentration in IDLN versus tumor tissue expression in tsEV-miR-34a-mimic group. We anticipated that this could be connected to the dual role of IL-10 in manipulating immune responses [60]. Along with the traditional tumor-promoting role of IL-10, recent studies provided evidence for the anti-tumor immune activating roles of IL-10 including elevation and activation of tumor-specific CD8T cells, higher expression of IFN-, and granzyme augmentation [61]. Application of pegylated IL-10 reinstitute CD8T cells in vivo, over-expressing IL-10 mice were protected from subsequent carcinogenesis, and on top of that, IL-10 deleted mice displayed poor immune surveillance toward tumor induction [62, 63]. In Martin oft study, it has been shown that IL-10 amplified the cytotoxicity of tumor-resident CD8 T by an induction in production of IFN-γ and granzyme [60] and on the other hand Segal et al. has proved that in a glioma model, the glioma specific CD4T cells that were isolated from spleen of tumor bearing mice produce a large amount of IL-10 after exposure to tumor lysate and manage tumor rejection [64].

In our study tumor specific CD4T cells in IDLN might be the major source of IL-10 production, in this regard the exact different roles and the source of IL-10-producing cells in the tumor and IDLN after tsEV-miR-34a-mimic therapy need to be highlighted in future studies.

According to our pathology results and considering all the possible changes detected in immune response and TME, the mice receiving mimic treatment had higher survival rate, and a clear difference in metastasis to the lungs, liver, and tumor-draining lymph nodes, compared to the other groups, which could be due to smaller size and weight of the tumor in them. The higher severity of necrosis observed in the mimic group compared to the other groups reflects the ability of this therapeutic agent to induce effective apoptosis.

Up to now, miR-34a-mimic replacement therapy has been tested in different studies. Baradaran et al. showed that a combination therapy of miR-34a and miR-200c is effective in controlling HIF which caused a block in VEGF and MMP9 expression in the breast cancer xenograft model [6]. In another study, the application of exosome-coated-miR-34a in Panc28-bearing mice showed great tumor inhibitory functions [65]. These studies just evaluated miR-34a effects from the tumor side and its immune-related effects were neglected.

In a recent study by Hart et al. [66, 67], some immune-suppressing effects were detected for miR-34a-mimic therapy in CD4T/CD8T cells which were in opposed to our results and T cell activating effects report of Zhong et al. [68]. We estimated that our immune beneficial results mainly originated from the route of injection and tsEV features. As stated by Seder et al., S.C. injection provided a local effector response whereas I.V. administration provided a systemic memory response [69]. Since we did not detect a powerful effect in splenocytes compared to IDLN-isolated lymphocytes, tsEV-miR-34a-mimic could just temporarily restrict tumor progression. Moreover, it seems that tumor suppressive effects of tsEV-miR-34a-mimic indirectly provided a better TME for immune cells.

In the tsEV-treated group some beneficial effects were also observed, including a reduction in tumor target genes (STAT3, PD-L1, VEGF, MMP9, and MMP2) and immune suppressor transcription factors (RORγc, FOXP3, and GATA3). The pattern of cytokine in the tsEV group was also skewed to the reduction of IL-6. TsEV induced partial tumor cytotoxicity compared to the control.

In a broader view, miR-34a-mimic could synergize the effects of tsEV therapy. As we have applied tsEVs from FBS-free cultured cells, they may contain massive amounts of biohazard molecules, owing to their recent possible roles in aiding autophagy- machinery [70], which becomes threatening to tumor-promoting plans under normal conditions. Additionally, tsEV could carry tumor-associated/specific antigens that could favorably activate tumor-specific CD8T cells and create an anti-tumor immune response [71–73].

The irregular-partial anti-tumor effects detected in NC-tsEV could be linked to two causes including its derivation from starved CT-26 cells or the nature of CT-26 itself which is a hot tumor and provides a completely new set of antigens for immune cells [74]. Further investigation to define the exact mechanism is needed in this field.

Although our study provides novel findings of application of tumor derived EVs for TNBC therapy, it was not free of limitations. The exact mechanism should be determined by examining other important cancer related immune responders such as myeloid derived suppressing cells, tumor associated macrophages, neutrophils and fibroblasts. We suggested these evaluations to have a complete view of immune response upon application of tsEV-miR-34a therapy in TNBC. The next possible step for this type of therapy might be the application of tsEVs from human sources such as MDA-MB231. Also, possible variations between different types of tumors based on their type of immunoactivity should not be neglected.

Conclusions

Based on what has been said so far, we suggest that tsEV could be an advantageous carrier in the context of MRT, especially for miR-34a. This study provided great evidence of the anti-tumor response of miR-34a from an immunological perspective instead of the tumor itself. tsEV-miR-34a mimic has the capability of changing TME and tumor cell’s behavior in favor of boosting immune response in TNBC. This platform could be considered as a complementary approach with other therapies of TNBC.

Abbreviations

tsEV

4T1-Tumor-derived small extracellular vesicle

tsEV-miR34a-mimic

MiR-34a- mimic loaded tsEV

tsEV-miR-34a- inhibitor

MiR-34a-inhibitor loaded tsEV

NC-miR-34a

Negative control – CT-26 derived tsEV

MRT

MicroRNA replacement therapy

Author contributions

M.H.: Conceptualization, Investigation, Formal analysis, Visualization, Writing-original draft, M. H.: Investigation, B.N.: Investigation, K.B.: Methodology, Writing- Review and Editing, N. M.: Methodology, Writing-Review and Editing, M.R.P: Investigation, D.A: Conceptualization, Supervision, Project Management, Writing- Review and Editing, Validation, Resources.

Funding

This research was financially supported by the Shahid Beheshti University of Medical Sciences under Grant [No.20846].

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval

This study was approved by the Institutional Ethical Committee and Research Advisory Committee of Shahid Beheshti University of Medical Sciences (ID: IR.SBMU.MSP.REC.1398.581), and experimentation was conducted based on the approved guidelines.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.